Re: [Qemu-devel] [PATCH v6 20/26] memory: Access MemoryRegion with endianness

[Paolo Bonzini]

On 07/08/19 19:49, Richard Henderson wrote:
> On 8/7/19 1:33 AM, address@hidden wrote:
>> @@ -551,6 +551,7 @@ void virtio_address_space_write(VirtIOPCIProxy *proxy, 
>> hwaddr addr,
>>          /* As length is under guest control, handle illegal values. */
>>          return;
>>      }
>> +    /* FIXME: memory_region_dispatch_write ignores MO_BSWAP.  */
>>      memory_region_dispatch_write(mr, addr, val, size_memop(len),
>>                                   MEMTXATTRS_UNSPECIFIED);
>>  }
> 
> Here is an example of where Paolo is quite right -- you cannot simply add 
> MO_TE
> via size_memop().  In patch 22 we see
> 
>> @@ -542,16 +542,15 @@ void virtio_address_space_write(VirtIOPCIProxy *proxy, 
>> hwaddr addr,
>>          val = pci_get_byte(buf);
>>          break;
>>      case 2:
>> -        val = cpu_to_le16(pci_get_word(buf));
>> +        val = pci_get_word(buf);
>>          break;
>>      case 4:
>> -        val = cpu_to_le32(pci_get_long(buf));
>> +        val = pci_get_long(buf);
>>          break;
>>      default:
>>          /* As length is under guest control, handle illegal values. */
>>          return;
>>      }
>> -    /* FIXME: memory_region_dispatch_write ignores MO_BSWAP.  */
>>      memory_region_dispatch_write(mr, addr, val, size_memop(len),
>>                                   MEMTXATTRS_UNSPECIFIED);
> 
> This is a little-endian store -- MO_LE not MO_TE.

Or leave the switch statement aside and request host endianness.  Either
is fine.

Paolo